Next Article in Journal
Patient Experience with a Gynecologic Oncology-Initiated Genetic Testing Model for Women with Tubo-Ovarian Cancer
Next Article in Special Issue
The Burden of Health-Related Out-of-Pocket Cancer Costs in Canada: A Case-Control Study Using Linked Data
Previous Article in Journal
Marrying Story with Science: The Impact of Outdated and Inconsistent Breast Cancer Screening Practices in Canada
Previous Article in Special Issue
The Health Economics of Metastatic Hormone-Sensitive and Non-Metastatic Castration-Resistant Prostate Cancer—A Systematic Literature Review with Application to the Canadian Context
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:

Cancer Premature Mortality Costs in Europe in 2020: A Comparison of the Human Capital Approach and the Friction Cost Approach

Paul Hanly
Marta Ortega-Ortega
2 and
Isabelle Soerjomataram
School of Business, National College of Ireland, Dublin 1, Ireland
Department of Applied Economics, Public Economics and Political Economy, Faculty of Economics and Business, Complutense University of Madrid, Campus de Somosaguas, 28223 Madrid, Spain
International Agency for Research on Cancer, 69372 Lyon, France
Author to whom correspondence should be addressed.
Curr. Oncol. 2022, 29(5), 3552-3564;
Submission received: 16 March 2022 / Revised: 1 May 2022 / Accepted: 9 May 2022 / Published: 13 May 2022


The inclusion of productivity costs can affect the outcome of cost-effectiveness analyses. We estimated the value of cancer premature mortality productivity costs for Europe in 2020 using the Human Capital Approach (HCA) and compared these to the Friction Cost Approach (FCA). Cancer mortality data were obtained from GLOBOCAN 2020 by sex and five-year age groups. Twenty-three cancer sites for 31 European countries were included. The HCA and the FCA were valued using average annual gross wages by sex and age group and applied to Years of Potential Productive Life Lost. 2020 friction periods were calculated and all costs were in 2020 euros. Estimated cancer premature mortality costs for Europe in 2020 were EUR 54.0 billion (HCA) and EUR 1.57 billion (FCA). The HCA/FCA cost ratio for Europe was 34.4, but considerable variation arose across countries (highest in Ireland: 64.5 v lowest in Czech Republic: 11.1). Both the HCA and the FCA ranked lung, breast and colorectal as the top three most costly cancers in Europe, but cost per death altered rankings substantially. Significant cost differences were observed following sensitivity analysis. Our study provides a unique perspective of the difference between HCA and FCA estimates of productivity costs by cancer site and country in Europe.

1. Introduction

Cancer is a leading cause of death globally, accounting for almost 10 million deaths in 2020, of which 20% arose in Europe [1]. This public health burden translates into a substantial economic and societal burden, which amounted to almost EUR 200 billion in Europe in 2018, consisting of both direct costs (cancer-specific health expenditure including expenditure on drugs) and indirect costs (premature mortality, morbidity and informal care costs) [2]. A considerable portion of the total cost burden of cancer is accounted for by productivity losses, of which premature mortality costs comprise the largest share.
In the vast majority of published cost of illness studies, productivity costs have been evaluated according to the Human Capital Approach (HCA) [3]; however, the Friction Cost Approach (FCA) offers an alternative [3,4]. The FCA tends to be more prevalent in Canada, Netherlands and Germany, which were early adopters and endorsers of including FCA estimated costs in their national pharmacoeconomic guidelines. The FCA is predicated on the assumption that unemployment levels above a frictional level exist in many economies, and so employees lost from work through illness and premature mortality can be replaced from a pool of the unemployed [5]. The productivity loss to the economy, and society at large, is therefore assumed to be transitory and smaller than the HCA equivalent cost, which measures productivity costs over the course of an entire working life cut short by premature death.
Most cancer-related premature mortality productivity cost studies to date apply the HCA to value productivity loss [6]. Indeed, only 2 European studies from a review of 17 studies included cancer mortality FCA-derived productivity costs, and even then, they were presented in addition to comparator HCA estimates [6]. More general reviews of the applied cost of illness literature have shown that productivity costs estimated according to the FCA are smaller than the equivalent HCA estimates across disease type, but that the relative magnitudes vary widely, and the divergence is greater for chronic illnesses [3]. Of the reviewed studies that focused on cancer mortality, HCA/FCA cost ratios ranged from 24 to 48 [3].
Although a number of studies in the literature have attempted to explore the differences between cancer-related productivity costs estimated according to the HCA and the FCA, these were limited to single, or just a few, cancer sites and located in a single country [7,8,9,10]. No previous study has attempted to estimate FCA derived productivity costs associated with cancer premature mortality in Europe across multiple sites using a standardized methodological approach to friction period estimation and directly compared these to the alternative HCA. Our study is therefore unique and aims to estimate the value of productivity losses due to cancer-related premature mortality across Europe in 2020 according to the HCA and FCA, and compare estimates by country, sex and cancer site.

2. Materials and Methods

2.1. Mortality Data and Approach

Cancer mortality data consisted of the number of deaths for each cancer site by sex and five-year age groups (from 15–64 years where deaths were assumed to occur at the midpoint of each 5-year age group) and was obtained from the Global Cancer Observatory, GLOBOCAN 2020 database (, accessed on 4 March 2022). Twenty-three individual cancer sites in addition to a total for all cancer sites were considered following the International Classification of Diseases, 10th revision (ICD-10). We included 31 European countries (the 27 countries of the European Union (EU), plus Norway, Switzerland, Iceland and the United Kingdom (UK)), which we hereafter refer to as ‘Europe,’ and four European regions (Northern, Southern, Central-Eastern, Western Europe).
Our estimates focused on paid production productivity loss, defined by Ortega-Ortega et al. [11] as the monetary value produced by a person who is working in the labour market. Therefore, we estimated the productivity lost in the productive age group, assumed to be between 15–64 years old considering the average statutory retirement in Europe at age 65 and in line with previous premature mortality cost studies in Europe [11,12].
Years of potential productive life lost (YPPLL) were calculated using the Standard Lifetable from the World Health Organisation (WHO) (, accessed on 3 March 2022), which reflects the years of potential production that have been lost due to each premature cancer death before 65 years of age by sex and country (YPPLL = life expectancy up to 65 − age at cancer death). YPPLL were subsequently multiplied by market annual gross wages by age group, sex and country to generate productivity costs according to two valuation methods.

2.1.1. Human Capital Approach (HCA)

The HCA is the method most commonly used in the literature to estimate labour productivity lost. It estimates the monetary value of a stream of output over a working lifetime that is cut short by premature death [13] and therefore is an estimate of the potential labour productivity lost when a person dies prematurely.
We valued productivity lost using average annual gross wages by sex, age group (15–29, 30–49, 50–59, 60–64 years) and country sourced from the Structure of Earnings Survey 2018 [13] and inflated to 2020 values using the Harmonized Index of Consumer Price [14], and accounted for country-specific macroeconomic conditions using national unemployment rates for 2020 [15] and labour force participation rates for 2020 [16], disaggregated by sex and age group (15–29, 30–39, 40–49, 50–59, 60–64 years). Present and future flows of potential productivity lost were valued by multiplying YPPLL by the sex-, age- and macroeconomic condition-adjusted wages for each country.
We assumed that wages grew at the average Gross Domestic Product (GDP) growth rate between 2000 to 2020 for each European country (base case) [17], and future costs were discounted at 3.5% per annum (base-case). All costs were expressed in 2020 euros.

2.1.2. Friction Cost Approach

The estimation of the FCA required an estimate of the friction period (the period of time to restore the production level of a firm following the loss of an employee and to fill a vacancy that has arisen) and monetary valuation data for lost production to the economy. Our estimate of country-specific friction periods, updated to 2020, followed a recently published methodology [18]. Initially, the average vacancy duration following the loss of an employee was calculated according to:
Annual   vacancy   duration   ( VD ) = 365     i = 1 4 V i 4 i = 1 4 M i
  • V = stock of unfilled vacancies
  • M = flows of filled vacancies
Estimates for the stock of unfilled vacancies were derived from annual Labour Force Surveys reported by Eurostat in 2020 [19]. Vacancy stock data are missing for France, Italy, Denmark, Hungary and Malta. In the case of Italy, Denmark, Hungary and Malta, we used average friction periods from each of their European regions and used these as approximate estimates of the friction period. For France, and following Hanly et al. [18], we used proxy estimates for vacancy stock (, accessed on 10 January 2022). Estimates of the annual flows of filled vacancies per country were calculated by multiplying the total number of occupied posts in each country [20] by the percentage of newly occupied posts in the last 12 months in that country [21]. Following previous studies, we added an additional 4 weeks to the country-specific vacancy duration periods [18], and we assumed an elasticity of annual labour time versus annual labour productivity of 1 [18].
Using the same YPPLL data as for the HCA, we valued cancer-related premature mortality by multiplying YPPLL by country-specific friction periods and by country-, sex- and age-specific average annual gross 2020 wages, following adjustments for country-, sex- and age-specific unemployment and labour force participation rates.

2.2. Sensitivity Analysis

Separate sensitivity analyses were conducted for the HCA and FCA. For the HCA, a sensitivity analysis was conducted using maximum and minimum country-specific GDP growth between 2000 and 2020 and 0% and 6% discount rates. For the FCA, a sensitivity analysis was run on maximum and minimum estimates of the friction period between 2008 and 2018 based on results reported by Hanly et al. [18] and on an estimate of the friction period for 2018, prior to the outbreak of COVID-19, for comparative purposes. Excel and Stata (version 18, StataCorp LLC, College Station, TX, USA) were used for analysis.

3. Results

3.1. Cancer Mortality and Years of Potential Productive Life Lost (YPPLL) in Europe

As shown in Table 1, a total of 336,964 premature cancer-related deaths (15–64 years of age) were estimated for 2020 across Europe, which amounted to 2,833,925 YPPLL. Of these deaths, Western Europe accounted for the largest portion (36%), followed by Southern Europe (25%), Central-Eastern Europe (24%) and Northern Europe (16%).

3.2. Total Premature Mortality Costs by Region and Country

In Table 1, the total estimated cancer premature mortality cost for Europe in 2020 was EUR 54.02 billion according to the HCA and EUR 1.57 billion according to the FCA, with HCA costs therefore 34.4 times the size of the equivalent FCA costs. Across the four European regions, this ratio varies from 29.7 in Western Europe (HCA = EUR 27.18 billion; FCA = EUR 0.92 billion) to 49.9 in Southern Europe (HCA = EUR 10.62 billion; FCA = EUR 0.21 billion).
From a country-specific perspective, the ranking of premature mortality costs remained similar between the HCA and FCA, with four (Germany, United Kingdom, France and Italy) of the top five ranked countries by cost burden being the same regardless of valuation approach employed (Spain ranks as fifth most burdensome by cancer cost in Europe according to the HCA, but eighth according to the FCA). Despite this degree of relative similarity in the ranking of the cancer cost burden by country, there was considerable variation in the HCA/FCA cost ratio across countries. The highest cost ratios were recorded for Ireland (64.5), Spain (63.5) and Portugal (60.0), while the lowest arose in Germany (24.3), Belgium (20.2) and Czechia (11.1).

3.3. Premature Mortality Costs per Cancer Death by Region and Country

The estimated cancer premature mortality cost per death for Europe in 2020 was EUR 160,318 according to the HCA and EUR 4656 according to the FCA (Table 1). Although the HCA/FCA cost ratios were the same as those for total premature mortality cost, the ranking of the cancer cost burden by country changed considerably. The top three costliest countries for cancer according to the HCA were Switzerland (EUR 456,884), Denmark (EUR 300,750) and Iceland (EUR 281,944). Only Switzerland (EUR 9276) remained in the top three costliest countries according to the FCA, with Belgium (EUR 10,387: 1st) and Germany (EUR 9221: 3rd) emerging as high-cost countries using this approach. Likewise, differences emerged in the ranking of the least costly countries with Croatia (EUR 50,441), Romania (EUR 49,208) and Bulgaria (EUR 46,526) ranked lowest according to the HCA, while Poland (EUR 1236), Lithuania (EUR 1138) and Bulgaria (EUR 988) were lowest ranked by the FCA.

3.4. Premature Mortality Costs by Gender, by Region and Country

The total estimated premature mortality cancer cost for Europe in 2020 was EUR 33.81 billion (EUR 176,787 per death) for males and EUR 20.20 billion (EUR 138,686 per death) for females according to the HCA (Table 2). The equivalent cost for the FCA was EUR 1.02 billion (EUR 5331 per death) for males and EUR 0.55 billion (EUR 3771 per death) for females, which resulted in a HCA/FCA cost ratio of 33.2 for male cancer costs and 36.8 for female cancer costs. By region, the highest HCA/FCA cost ratio arose in Southern Europe, where HCA costs were 48.3 times higher for males and 53.0 times higher for females, while the lowest ratio occurred in Western Europe where HCA costs were 28.7 times higher for males and 31.7 times higher for females. This higher HCA/FCA cost ratio for females occurred across Europe where application of the FCA appears to have exacerbated the gender disparity in the valuation of the cancer burden. Whereas the male/female ratio for cancer deaths was 1.31, the application of the HCA increased this ratio to 1.7, and the application of the FCA further widened the male/female burden gap to 1.9. Therefore, in countries such as Portugal (2.5), Greece (2.4) and Spain (2.2), the male/female FCA cost ratio rose above 2 even though the male/female ratio of cancer death was below 2 (Portugal: 1.8, Greece: 1.4, Spain: 1.5).

3.5. Premature Mortality Costs by Cancer Site

In Table 3, both the HCA and the FCA valuation approaches ranked lung (HCA: EUR 11.08 billion; FCA: EUR 0.39 billion), breast (HCA: EUR 5.04 billion; FCA: EUR 0.12 billion), colorectum (HCA: EUR 4.85 billion; FCA: EUR 0.14 billion), brain (HCA: EUR 4.2 billion; FCA: EUR 0.09) and pancreatic (HCA: EUR 3.2 billion; FCA: EUR 0.11 billion) cancer as the five most costly according to total premature mortality costs. Together, these five cancers accounted for 53% (HCA) and 54% (FCA) of the total cancer cost burden.
Cancer premature mortality costs per death according to the HCA and FCA revealed an alternative perspective on the cancer burden. According to this metric, the most costly cancers as valued by the HCA were Hodgkin Lymphoma (EUR 298,093), melanoma of the skin (EUR 252,204), brain (EUR 247,090), leukaemia (EUR 207,641) and non-Hodgkin Lymphoma (EUR 188,987). Each of these five cancers are associated with relatively high HCA/FCA cost ratios above 40, with Hodgkin Lymphoma in particular showing a cost ratio of 67.6 (the highest across all cancer sites). This is, in part, a consequence of the younger age of death associated with these cancers. The FCA included only two cancers from this list (brain (EUR 5304) and melanoma of the skin (EUR 5461)) and added to the top five most costly cancers per death head and neck (EUR 4766), kidney (EUR 5040) and oesophageal (EUR 6039) cancer.

3.6. Sensitivity Analysis

For the HCA (Table 4), substantial differences in the total value of lost productivity were observed following the application of minimum and maximum national economic growth rates as measured by GDP average changes. Minimum growth rates reduced costs by 22% on average compared to the base case, with a reduction of 37% in the case of Central-Eastern Europe. Costs increased by 23% on average when maximum growth rates were used with Central-Eastern Europe, again showing the highest divergence from the base case with an increase of 63%. Altering the discount rate also resulted in considerable changes, with a 0% discount rate increasing costs by 24% compared to the base case and a 6% discount rate reducing costs by 12%.
For the FCA (Table 5), substantial differences in estimated productivity costs by region and for Europe as a whole were observed following the application of country-specific minimum and maximum friction periods from the period 2008–2018. Productivity costs increased by 22% in the case of maximum friction periods for Europe. This altered the HCA/FCA cost ratio from 34 in the base case to 28. Application of the minimum friction period from 2008–2018 resulted in a decrease in productivity costs by 15% across Europe, with a resultant increase in the HCA/FCA ratio to 41. Estimated productivity costs using 2018 friction periods increased over the base case cost in line with expectations given the assumed negative relationship between vacancy duration (and hence friction period) and the unemployment rate [18], and low unemployment rates across Europe pre COVID-19 pandemic.

4. Discussion

The dominant valuation approach for productivity costs in the literature is the HCA; however, criticisms abound of this approach [4,22,23,24], and primary amongst these is the overestimation of productivity costs due to the assumptions that employees are not replaced following illness or death [25]. Although it has been suggested that the FCA generates more realistic estimates of productivity costs than the HCA [5], issues with regard to a standardised approach to friction period estimation across countries, and a lack of data necessary to compute accurate friction periods by country remain. It is this issue that the current study confronts by applying a standardised methodological framework for friction period estimation across Europe to estimate comparable multi-country FCA productivity cost estimates for cancer.
Our study revealed that cancer premature mortality costs for Europe were EUR 54.02 billion (HCA) and EUR 1.57 billion (FCA), indicating that HCA costs are 34 times larger than FCA costs. Estimates of the HCA/FCA costs ranged from 29.7 in Western Europe to 49.9 in Southern Europe by region, from 11 in Czechia to 65 in Ireland by country, and from 21 for prostate cancer to 67 for Hodgkin Lymphoma by cancer site. Previous reviews of FCA and HCA cancer productivity costs reveal similar variability in their cost ratios. Pike et al. [3] found that cancer studies that included mortality costs had HCA/FCA ratios between 24 and 48. By cancer site, previous single-country studies have shown HCA/FCA ratios of 49 and 73 for breast cancer in Spain [9] and Ireland [7], respectively (42 for breast cancer in our study), 56 for prostate cancer in Ireland (21 for prostate cancer in our study) [7], 55 for brain cancer in Spain (47 for brain cancer in our study) [9]) and 47 for haematologic cancer in Spain (46 in our study) [10]. Therefore, while results vary in terms of the magnitude of the HCA/FCA ratio, high-fatality diseases such as cancer do appear to generate HCA cost estimates considerably in excess of their FCA equivalents, and this result is substantiated by our estimates of cancer in Europe.
The drivers of the variability in the HCA/FCA cost ratios are varied and nuanced and arise due to a combination of underlying epidemiological features and macroeconomic features including labour market data used for valuation purposes. For example, in the case of prostate cancer where the HCA/FCA ratio is 21, the mean age of death is 59.4, which is the oldest across all cancers (average age of death for all cancers = 55.6). Consequently, there are fewer remaining productive years to cost according to the HCA diminishing the gap between the HCA and the shorter FCA costing method. Similarly, Hodgkin Lymphoma, with an HCA/FCA cost ratio of 67, had the youngest average age of death across all cancers at 46.8, allowing HCA costs to accumulate over a longer time frame. In contrast, by country, the key drivers of the low HCA/FCA cost ratio for Czechia (11), for example, were more economic in nature. Czechia had the longest friction period in the study (247 days in 2020 versus 74 days European average) and therefore had the potential to generate higher productivity costs according to the FCA. This was primarily driven by the size of the vacancy rate in Czechia (5.3% in 2020; the highest value across all EU countries), which translated into a comparatively large job vacancy stock figure. Ireland, which produced the highest HCA/FCA cost ratio of 65, had the fourth lowest friction period in Europe for 2020 (47 days), the fifth lowest age of cancer death on average (55.0 years old versus average in Europe 55.7) and, consequently, one of the highest YPPLL per number of deaths ratios in Europe (8.4, sixth highest in Europe). This range of economic and epidemiological factors indicates the complexity of factors driving productivity cost variation between both valuation methods.
Interestingly, despite the variability in HCA/FCA cost ratios highlighted across Europe, a key finding of this study was that the ranking of total premature mortality cost burden by region, country and cancer site remained generally independent of the selected valuation approach. For example, four of the top five ranked countries by total cancer productivity cost burden were the same according to both HCA and FCA, and, by cancer site, the top five ranked cancers by total cost burden were the same for both valuation approaches. This was not the case for cancer premature mortality cost per death though, where the ranking of cancer burden changed considerably by region and by cancer site depending on selected valuation approach. Traditional macroeconomic factors such as high average gross wages and low unemployment rates explain the ranking of cancer productivity cost burden by country according to the HCA (first: Switzerland, second: Denmark and third: Iceland), while the FCA highlights the speed at which an employee is replaced after leaving the workplace and ranks countries such as Belgium (first), Germany (third) and Czechia (fifth) as high cost based on the length of their friction periods (in addition to high wages and low unemployment levels in the case of Belgium and Germany).
Per death productivity costs by cancer site, on the other hand, seem to be driven by the average age of death for both the HCA and FCA with melanoma skin cancer (average age of death = 52.2) and brain cancer (average age of death = 52.2), both ranking in the top five most costly cancers according to both valuation approaches. Additionally, in the case of the FCA, the ratio of male deaths to female deaths per cancer impacts on estimated costs. Oesophageal cancer (first) and head and neck cancer (fifth) emerge as high cost per death cancers in spite of their absence from the top five most costly as measured by the HCA. Both have a male/female cancer death ratio far in excess of the average across all cancers (5.0 and 4.8, respectively, versus an average European ratio of 1.3). This represents a significant finding, and while we know that estimating economic costs can offer an interesting alternative public health perspective on the burden of diseases, we additionally highlight that the choice of valuation lens can considerably alter the reported burden of high fatality illnesses such as cancer when costs are expressed on a per death basis.
Interesting findings also emerge following an examination of the gender aspects of the differences between the HCA and the FCA. Our findings show that the FCA intensified the gender disparity in the valuation of the cancer burden. Whereas previous studies [2,11,12] have revealed the degree to which male cancer premature mortality costs exceed female costs following the application of market wages through the HCA, our results reveal how the male/female cost ratios increased on average from 1.7 for the HCA to 1.9 for the FCA. Future researchers and those implementing economic evaluations from a societal perspective should be cognisant of this gender effect and could perhaps look to use national average gross wages rather than gender specific wages to overcome this.
Following the results of previous studies, our sensitivity analysis highlighted the key role played by the length of a friction period in FCA derived productivity costs and how changes in this variable over time due to underlying macroeconomic conditions can impact productivity costs. FCA theory predicts shorter friction periods as unemployment rises [26]. For example, due to the macroeconomic effects of the COVID-19 pandemic, European unemployment rates rose considerably between 2018 and 2020, resulting in shorter friction periods in 2020 compared to 2018. The outcome of this is that the FCA productivity costs of cancer are 16% higher based on labour market conditions from the pre-COVID 2018 period compared to the COVID-19-impacted 2020 estimates. Our analyses further show that the HCA/FCA ratio increased to 41 as a consequence of using country-specific minimum friction period estimates from 2008–2018 and fell to 28 following the use of maximum estimates over this time frame. This study therefore highlights the dependence of derived FCA productivity estimates on the extant labour market conditions of a country at a point in time and the importance of updating these regularly on a national basis to obtain accurate productivity estimates over time.
Our study highlights the benefits of using a standardised methodological framework for friction period estimation, offering the potential to compare derived costs to the HCA. The findings offer valuable lessons for researchers in the range of, and discrete nature of, the drivers of costs for each approach and will aid in informing researcher and policy-maker choice between valuation approaches by providing transparency in the costing process and providing a practical example as to how costs diverge for a key constituent of indirect costs over the course of a working lifetime.
Our results also feed into the literature on cost-effectiveness analysis. Previous studies have shown that the inclusion of productivity costs in cost-effectiveness analyses have the potential to significantly impact the outcome of the analysis [27]. Recent reviews have shown that productivity cost inclusion in economic evaluations generally leads a more favourable cost-effectiveness outcome and, in a significant minority of cases, changes the incremental cost-effectiveness ratio (ICER) from positive to negative, transforming the new treatment into a cost-saving intervention [27,28]. Nevertheless, most pharmacoeconomic guidelines do not advocate a societal perspective for economic evaluations, and therefore, it is left to researcher discretion in many cases to include them. This problem is exacerbated by the lack of a universally recognized framework for calculating productivity costs; consequently, valuation approaches vary widely [4,27]. The use of a standardized methodological framework in our study to estimate FCA productivity costs based on publicly available national data sources provides a potential avenue to overcome some of the current variability noted in valuation approaches, hence strengthening the argument for their inclusion in future cost-effectiveness analyses.
Nevertheless, there are several limitations of our study. Our study is an underestimate of the productivity costs of cancer in Europe due to the explicit exclusion of other forms of cancer-related lost productivity including morbidity in the form of temporary work absences, permanent absenteeism, reduced working hours and presenteeism. The lack of a standardised database across Europe for these productivity costs for cancer resulted in their exclusion in this case. We also excluded unpaid productivity costs from our study, which have been shown to be significant for cancer [11]. This was a deliberate choice made by the authors, as non-market activity, by its nature, does not result in the need to replace employees in a firm. To aid the comparability of our estimates with previous studies of cancer productivity costs in Europe, we also excluded those individuals who die prematurely from cancer but would have worked beyond 65 years of age. Robust estimates for individuals working beyond 65 are difficult due to the limited economic data availability for wage rates, labour force participation rates and unemployment rates for this group. Our study employed gross national wage rates rather than self-reported wage rates and therefore does not reflect variations in the socio-economic background of the population under study. Nor do our estimates reflect variations in vacancy duration by occupation due to a lack of routinely collected data at a national level on vacancy stock data by sector or by occupation. Our forecasted labour market data used for the HCA was based on an average of historic trends, which may not accurately reflect future trends. This is a common shortcoming of HCA estimates, however, and one we undertook sensitivity analysis to mitigate as much as possible. Given the 2020 base case year in this study, the labour market data used for valuation purposes has been impacted the COVID-19 pandemic. We have undertaken broad ranging sensitivity analysis as a consequence by modelling the impact of future wage growth and friction period estimates on our results, and we have discussed how our FCA costs are likely influenced by these atypical macroeconomic conditions.

5. Conclusions

Our study provides a unique perspective on the difference between HCA and FCA estimates of cancer-related premature mortality productivity costs in Europe using a standardized methodological approach to friction period estimation. We highlight the magnitude of the difference between HCA and FCA estimates for cancer and, importantly, draw attention to the dependency of the FCA on the extant macroeconomic and labour market conditions of a country. We offer insights for researchers and policy makers into the key reasons for the differences in costs between valuation approaches while providing transparency on their application in practice.

Author Contributions

Conceptualization, P.H., M.O.-O. and I.S.; Methodology, P.H., M.O.-O. and I.S.; Software, P.H. and M.O.-O.; Validation, P.H. and M.O.-O.; Formal Analysis, P.H. and M.O.-O.; Investigation, P.H., M.O.-O. and I.S.; Resources, P.H., M.O.-O. and I.S.; Data Curation, P.H., M.O.-O. and I.S. Writing—Original Draft Preparation, P.H. and M.O.-O.; Writing—Review and Editing, P.H., M.O.-O. and I.S.; Visualization, P.H., M.O.-O. and I.S.; Supervision, P.H.; Project Administration, P.H; Funding Acquisition. All authors have read and agreed to the published version of the manuscript.


This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Publicly available datasets were analysed in this study. These data can be found here: (accessed on 7 January 2022) and (accessed on 8 January 2022).

Conflicts of Interest

The authors declare no conflict of interest.


  1. Ferlay, J.; Ervik, M.; Lam, F.; Colombet, M.; Mery, L.; Piñeros, M. Global Cancer Observatory: Cancer Today; International Agency for Research on Cancer: Lyon, France, 2020; Available online: (accessed on 15 February 2022).
  2. Hofmarcher, T.; Lindgren, P.; Wilking, N.; Jönsson, B. The cost of cancer in Europe 2018. Eur. J. Cancer 2020, 129, 41–49. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  3. Pike, J.; Grosse, S.D. Friction Cost Estimates of Productivity Costs in Cost-of-Illness Studies in Comparison with Human Capital Estimates: A Review. Appl. Health Econ. Health Policy 2018, 16, 765–778. [Google Scholar] [CrossRef] [PubMed]
  4. Krol, M.; Brouwer, W.; Rutten, F. Productivity costs in economic evaluations: Past, present, future. Pharmacoeconomics 2013, 31, 537–549. [Google Scholar] [CrossRef] [PubMed]
  5. Kigozi, J.; Jowett, S.; Lewis, M.; Barton, P.; Coast, J. Estimating productivity costs using the friction cost approach in practice: A systematic review. Eur. J. Health Econ. 2016, 17, 31–44. [Google Scholar] [CrossRef] [PubMed]
  6. Hanly, P.; Pearce, A.; Sharp, L. The cost of premature cancer-related mortality: A review and assessment of the evidence. Expert Rev. Pharmacoecon. Outcomes Res. 2014, 14, 355–377. [Google Scholar] [CrossRef] [PubMed]
  7. Hanly, P.; Timmons, A.; Walsh, P.M.; Sharp, L. Breast and prostate cancer productivity costs: A comparison of the human capital approach and the friction cost approach. Value Health 2012, 15, 429–436. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  8. Pearce, A.M.; Hanly, P.; Timmons, A.; Walsh, P.M.; O’Neill, C.; O’Sullivan, E.; Gooberman-Hill, R.; Thomas, A.A.; Gallagher, P.; Sharp, L. Productivity Losses Associated with Head and Neck Cancer Using the Human Capital and Friction Cost Approaches. Appl. Health Econ. Health Policy 2015, 13, 359–367. [Google Scholar] [CrossRef] [PubMed]
  9. Oliva, J.; Lobo, F.; López-Bastida, J.; Zozaya, N.; Romay, R. Indirect costs of cervical and breast cancers in Spain. Eur. J. Health Econ. 2005, 6, 309–313. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  10. Ortega-Ortega, M.; Oliva-Moreno, J.; Jiménez-Aguilera, J.D.D.; Romero-Aguilar, A.; Espigado-Tocino, I. Productivity loss due to premature mortality caused by blood cancer: A study based on patients undergoing stem cell transplantation. Gac. Sanit. 2015, 29, 178–183. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  11. Ortega-Ortega, M.; Hanly, P.; Pearce, A.; Soerjomataram, I.; Sharp, L. Paid and unpaid productivity losses due to premature mortality from cancer in Europe in 2018. Int. J. Cancer 2022, 150, 580–593. [Google Scholar] [CrossRef] [PubMed]
  12. Hanly, P.; Soerjomataram, I.; Sharp, L. Measuring the societal burden of cancer: The cost of lost productivity due to premature cancer-related mortality in Europe. Int. J. Cancer 2015, 136, E136–E145. [Google Scholar] [CrossRef] [PubMed]
  13. Eurostat. Mean Annual Earnings by Sex, Age and Economic Activity (earn_ses18_27). 2018. Available online: (accessed on 12 January 2022).
  14. Eurostat. HICP—Annual Data (Average Index and Rate of Change) (prc_hicp_aind). 2022. Available online: (accessed on 13 January 2022).
  15. Eurostat. Unemployment by Sex and Age—Annual Data (une_rt_a). 2022. Available online: (accessed on 12 January 2022).
  16. Eurostat. Active Population by Sex, Age and Citizenship (1 000) (lfsa_agan). 2022. Available online: (accessed on 12 January 2022).
  17. Eurostat. GDP and Main Components (Output, Expenditure and Income) [NAMA_10_GDP$DEFAULTVIEW]. 2022. Available online: (accessed on 20 January 2022).
  18. Hanly, P.; Ortega Ortega, M.; Pearce, A.; Soerjomataram, I.; Sharp, L. Advances in the methodological approach to friction period estimation: A European perspective. Soc. Sci. Med. 2020, 264, 113289. [Google Scholar] [CrossRef] [PubMed]
  19. Eurostat. Job Vacancies in Number and %—NACE Rev. 2, B-S, Quarterly Data [TPS00172__custom_2179636]. 2022. Available online: (accessed on 14 January 2022).
  20. Eurostat. Job Vacancy Statistics by NACE Rev. 2 Activity, Occupation and NUTS 2 Regions—Quarterly Data [JVS_Q_ISCO_R2__custom_2179956]. 2022. Available online: (accessed on 14 January 2022).
  21. Eurostat. Newly Employed [TESEM200]. 2022. Available online: (accessed on 14 January 2022).
  22. Johannesson, M.; Karlsson, G. The friction cost method: A comment. J. Health Econ. 1997, 16, 249–255. [Google Scholar] [CrossRef]
  23. Liljas, B. How to calculate indirect costs in economic evaluations. Pharmacoeconomics 1998, 13, 1–7. [Google Scholar] [CrossRef] [PubMed]
  24. Tranmer, J.E.; Guerriere, D.N.; Ungar, W.J.; Coyte, P.C. Valuing patient and caregiver time: A review of the literature. Pharmacoeconomics 2005, 23, 449–459. [Google Scholar] [CrossRef] [PubMed]
  25. Koopmanschap, M.A.; Rutten, F.F.; van Ineveld, B.M.; van Roijen, L. The friction cost method for measuring indirect costs of disease. J. Health Econ. 1995, 14, 171–189. [Google Scholar] [CrossRef]
  26. Hanly, P.; Koopmanschap, M.; Sharp, L. Valuing productivity costs in a changing macroeconomic environment: The estimation of colorectal cancer productivity costs using the friction cost approach. Eur. J. Health Econ. 2016, 17, 553–561. [Google Scholar] [CrossRef] [PubMed]
  27. Yuasa, A.; Yonemoto, N.; LoPresti, M.; Ikeda, S. Use of Productivity Loss/Gain in Cost-Effectiveness Analyses for Drugs: A Systematic Review. Pharmacoeconomics 2021, 39, 81–97. [Google Scholar] [CrossRef] [PubMed]
  28. Krol, M.; Papenburg, J.; Tan, S.S.; Brouwer, W.; Hakkaart, L. A noticeable difference? Productivity costs related to paid and unpaid work in economic evaluations on expensive drugs. Eur. J. Health Econ. 2016, 17, 391–402. [Google Scholar] [CrossRef] [PubMed] [Green Version]
Table 1. Number of premature deaths due to cancer * (15–65 years), age of death, Years of Potential Productive Life Lost (YPPLL), cancer-related premature mortality costs (PMC) (EUR, total and per death) by Human Capital Approach (HCA) and Friction Cost Approach (FCA) and HCA/FCA ratios for Europe by region and country in 2020.
Table 1. Number of premature deaths due to cancer * (15–65 years), age of death, Years of Potential Productive Life Lost (YPPLL), cancer-related premature mortality costs (PMC) (EUR, total and per death) by Human Capital Approach (HCA) and Friction Cost Approach (FCA) and HCA/FCA ratios for Europe by region and country in 2020.
Region/CountryDeathsAverage Age at DeathYPPLLHCA PMC
(EUR Millions)
(EUR Millions)
HCA PMC per DeathFCA PMC per DeathHCA/FCA Ratio
Central-Eastern Europe80,97055.7676,240504516262,358200131.1
Northern Europe51,59255.5442,34911,165280216,377541939.9
United Kingdom31,90155.2279,6636960177218,175555139.3
Southern Europe80,03255.4692,80110,620213132,682266049.9
Western Europe124,37055.81,022,53527,184915218,624735329.7
The Netherlands10,34856.181,347257089248,263857629.0
EUROPE-27+ Norway, Switzerland, Iceland, UK336,96455.62,833,92554,0151569160,318465634.4
* Cancer totals include non-melanoma skin cancer.
Table 2. Cancer *-related premature mortality costs (PMC) (EUR, total and per cancer death) by Human Capital Approach (HCA) and Friction Cost Approach (FCA) for Europe by region and country and sex in 2020.
Table 2. Cancer *-related premature mortality costs (PMC) (EUR, total and per cancer death) by Human Capital Approach (HCA) and Friction Cost Approach (FCA) for Europe by region and country and sex in 2020.
Males (M)Females (F)Sex Ratios
RegionHCA PMC (EUR Millions)FCA PMC (EUR Millions)HCA/FCA Ratio for PMCHCA PMC per DeathFCA PMC per DeathHCA PMC (EUR Millions)FCA PMC (EUR Millions)HCA/FCA Ratio for PMCHCA PMC per DeathFCA PMC per DeathM/F Deaths RatioM/F YPPLL RatioM/F HCA PMC RatioM/F FCA PMC Ratio
Central-Eastern Europe320810729.967,387225218375533.655,15916421.
Northern Europe640516738.3240,8546297476111242.4190,35644851.
United Kingdom402010737.5248,862664129407042.1186,71444341.
Southern Europe691814348.3150,352311037027053.0108,78120511.
Western Europe17,28360228.7243,3448473990131331.7185,69058621.
The Netherlands15205527.6294,70610,65710503431.1202,01765031.
EUROPE-27+ Norway, Switzerland, Iceland, UK33,814102033.2176,787533120,20054936.8138,68637711.
* Cancer totals include non-melanoma skin cancer.
Table 3. Cancer-related premature mortality costs (PMC) (EUR, total and per cancer death) by Human Capital Approach (HCA) and Friction Cost Approach (FCA) for Europe by cancer site in 2020.
Table 3. Cancer-related premature mortality costs (PMC) (EUR, total and per cancer death) by Human Capital Approach (HCA) and Friction Cost Approach (FCA) for Europe by cancer site in 2020.
Cancer SiteDeathsAverage Age at DeathYPPLLHCA PMC (EUR Millions)FCA PMC (EUR Millions)HCA PMC per DeathFCA PMC per DeathHCA/FCA Ratio
Head and neck14,60555.5128,508250872165,007476634.6
Melanoma skin591452.270,105149132252,204546146.2
Cervix Uteri718751.490,653124025172,315354548.7
Corpus Uteri373357.623,9853071182,197290128.4
Brain and Central Nervous System17,04152.2201,675421090247,090530446.6
Hodgkin Lymphoma88546.815,3382664298,093443967.6
Non-Hodgkin Lymphoma693854.268,148131133188,987473039.9
Multiple Myeloma385257.724,13148217125,171452127.7
All cancers336,96455.62,833,92554,0001569160,318465634.4
All cancers except non-melanoma skin336,07055.62,826,31553,9001565160,293465634.4
Table 4. Sensitivity analysis of cancer-related premature mortality costs (PMC) (EUR, total) using the Human Capital Approach with comparison to the base case *.
Table 4. Sensitivity analysis of cancer-related premature mortality costs (PMC) (EUR, total) using the Human Capital Approach with comparison to the base case *.
RegionHCA PMC (EUR Millions)% Change from BC
EUROPE. Discount Rate 0%67,10024
Central-Eastern Europe635026
Northern Europe14,10026
Southern Europe13,20025
Western Europe33,50023
EUROPE. Discount Rate 6%47,600−12
Central-Eastern Europe4410−13
Northern Europe9770−13
Southern Europe9360−12
Western Europe24,100−11
EUROPE. Minimum GDP growth of each country41,900−22
Central-Eastern Europe3180−37
Northern Europe7850−30
Southern Europe8090−24
Western Europe22,800−16
EUROPE. Maximum GDP growth of each country66,60023
Central-Eastern Europe825063
Northern Europe16,40046
Southern Europe12,10014
Western Europe29,90010
* Base case: up to 65 years old, Discount Rate = 3.5%, average Gross Domestic Product (GDP) growth of each country.
Table 5. Sensitivity analysis of cancer-related premature mortality costs (PMC) (EUR, total) using the Friction Cost Approach with comparison to the base case *.
Table 5. Sensitivity analysis of cancer-related premature mortality costs (PMC) (EUR, total) using the Friction Cost Approach with comparison to the base case *.
Region/CountryFriction Periods in DaysFCA PMC (EUR Millions)% Change from the Base CaseHCA/FCA Ratio
EUROPE. Friction period = 20187918231630
Central-Eastern Europe105.2173729
Northern Europe69.23422233
Southern Europe61.1225647
Western Europe91.710831825
EUROPE. Minimum friction periods (2008–2018)55.81331−1541
Central-Eastern Europe60.8100−3850
Northern Europe54.6265−542
Southern Europe44.9204−452
Western Europe66.7762−1736
EUROPE. Maximum friction periods (2008–2018)85.919202228
Central-Eastern Europe106.4174829
Northern Europe74.23462432
Southern Europe73.12531942
Western Europe92.511462524
* Base case: Friction periods in 2020 for each country.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Hanly, P.; Ortega-Ortega, M.; Soerjomataram, I. Cancer Premature Mortality Costs in Europe in 2020: A Comparison of the Human Capital Approach and the Friction Cost Approach. Curr. Oncol. 2022, 29, 3552-3564.

AMA Style

Hanly P, Ortega-Ortega M, Soerjomataram I. Cancer Premature Mortality Costs in Europe in 2020: A Comparison of the Human Capital Approach and the Friction Cost Approach. Current Oncology. 2022; 29(5):3552-3564.

Chicago/Turabian Style

Hanly, Paul, Marta Ortega-Ortega, and Isabelle Soerjomataram. 2022. "Cancer Premature Mortality Costs in Europe in 2020: A Comparison of the Human Capital Approach and the Friction Cost Approach" Current Oncology 29, no. 5: 3552-3564.

Article Metrics

Back to TopTop